From: George Guo <dongtai.guo@linux.dev>
To: chenhuacai@kernel.org, kernel@xen0n.name
Cc: rppt@kernel.org, liukexin@kylinos.cn, guodongtai@kylinos.cn,
loongarch@lists.linux.dev, linux-mm@kvack.org,
linux-kernel@vger.kernel.org
Subject: [PATCH] LoongArch: Allow memblock arrays to resize
Date: Thu, 19 Nov 2026 19:02:16 +0900 [thread overview]
Message-ID: <20261119100216.10182-1-dongtai.guo@linux.dev> (raw)
From: George Guo <guodongtai@kylinos.cn>
LoongArch never calls memblock_allow_resize(), so the memblock region
arrays cannot grow beyond their static sizes. For the reserved array
that is INIT_MEMBLOCK_RESERVED_REGIONS (INIT_MEMBLOCK_REGIONS +
NR_CPUS) entries. Once it is full, the next reservation that needs a
new region panics:
Kernel panic - not syncing: memblock: cannot resize reserved array
The memory setup code was derived from MIPS, which never enables
resizing either. Nothing on LoongArch needs the arrays to stay static.
When an array is full, memblock allocates a bigger one from free
memory, so resizing is only safe once all memory in use is reserved.
By the time setup_arch() calls arch_mem_init(), the kernel image, the
initrd, the firmware memory map, the device tree reserved-memory
regions and the crash kernel are reserved, so enable resizing there.
Co-developed-by: Kexin Liu <liukexin@kylinos.cn>
Signed-off-by: Kexin Liu <liukexin@kylinos.cn>
Signed-off-by: George Guo <guodongtai@kylinos.cn>
---
This was patch 5/8 of the LoongArch KHO series v6 [1]. Mike suggested
fixing it independently of KHO, and Huacai preferred arch_mem_init()
to platform_init() [1].
KHO for LoongArch needs this: after kexec, kho_memory_init() reserves
every preserved range, and luo_stress_files with 500 files overflows
the reserved array without CONFIG_SMP. A bigger array cannot land on
preserved memory, because KHO limits memblock allocations to its
bootmem areas by then.
Changes since v6 5/8:
- Sent on its own, without the KHO series.
- Reworded the commit message and the comment so they do not depend
on KHO.
[1] https://lore.kernel.org/all/20261029172956.20791-6-dongtai.guo@linux.dev/
arch/loongarch/kernel/setup.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/arch/loongarch/kernel/setup.c b/arch/loongarch/kernel/setup.c
index 6fa4a22a58fd..8c33346a123a 100644
--- a/arch/loongarch/kernel/setup.c
+++ b/arch/loongarch/kernel/setup.c
@@ -402,6 +402,9 @@ static void __init arch_mem_init(char **cmdline_p)
check_kernel_sections_mem();
+ /* All memory in use is reserved now, so memblock can grow its arrays. */
+ memblock_allow_resize();
+
memblock_set_bottom_up(true);
swiotlb_init(true, SWIOTLB_VERBOSE);
base-commit: 422e532809572ba1043192bf6a0c653bc68b3140
--
2.53.0
next reply other threads:[~2026-11-19 10:02 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-19 10:02 George Guo [this message]
2026-11-19 11:07 ` Mike Rapoport
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20261119100216.10182-1-dongtai.guo@linux.dev \
--to=dongtai.guo@linux.dev \
--cc=chenhuacai@kernel.org \
--cc=guodongtai@kylinos.cn \
--cc=kernel@xen0n.name \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=liukexin@kylinos.cn \
--cc=loongarch@lists.linux.dev \
--cc=rppt@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®